Well before the middle of the first year of life when canonical babbling begins, the human infant vocalizes in ways that manifest the emergence of critical capabilities for speech. Important aspects of these patterns of vocalization appear to occur rarely if ever in any non-human primate. Yet the patterns of usage of individual precanonical vocalization types have not been clearly delineated and quantified. For example, while some infant sounds (or features of sounds) appear to be used with fairly consistent functions, other identifiable precanonical sounds appear to be produced with neutral affect by infants in patterns of repetition of particular sounds and in patterns of systematic alternation among the sounds when infants are comfortably alone. The very same sounds appear to be used expressively with either negative or positive affect at other times. Further, the same sounds may on other occasions be directed to adults in interaction as indicated by eye contact accompanying the vocalizations. This sort of freedom to produce particular sounds in a variety of contexts appears to indicate the emergence of a primary foundation for speech. Still, science has provided no quantified characterization of the emergence of this flexibility of vocalization, nor of the ways that infants systematically communicate states and intentions, in spite of the vocal flexibility, through consistent usage of other more stable sounds or features of sounds in reliably identifiable contexts of facial affect, gesture, and/or interactive sequence. It is no longer beyond our capability to bridge this chasm in our understanding of precanonical infant vocalizations and their role in the foundations of speech. What remains is to fill the empirical void through intensive acoustic, auditory and visual analysis of vocalizations and the rich context in which they occur based on longitudinal recordings from human infants with intensive sampling and acoustic analysis on a small number of infants with a primary focus on the period from 2 to 7 months of age. The work is designed to evaluate how vocal communication in early infancy is accomplished through consistent usage of some sounds in context, and at the same time to assess ways that the flexible usage of a growing vocal repertoire in infancy can form a basis for speech. To the extent that precanonical category usage can be quantified, it may lay the foundations for early identification of disorders of communication.